Efficient Bluetooth receiving end monaural upmixing method and device, medium and equipment

Abstract

The invention discloses an efficient Bluetooth receiving end monaural upmixing method and device based on non-negative matrix factorization, a storage medium and equipment, and belongs to the technical field of Bluetooth audios, the method comprises the following steps: at a Bluetooth receiving end, decoding an input monaural AAC code stream or a dual-channel code stream to time domain noise shaping, obtaining a spectral coefficient and a corresponding amplitude spectral coefficient; performing monaural judgment on the spectral coefficients, and if the judgment result is monaural, constructing a corresponding amplitude spectral coefficient matrix according to the type of the spectral coefficients; performing non-negative matrix factorization on the amplitude spectrum coefficient matrix to obtain a corresponding basis matrix and an activation matrix; calculating a main sound source spectral coefficient and an environment sound spectrum coefficient according to the spectral coefficient, the amplitude spectral coefficient, the basis matrix and the activation matrix; and continuing to decode the main sound source spectrum coefficient and the environmental sound spectrum coefficient, and outputting main sound source PCM data and environmental sound PCM data. According to the invention, the environment sound is generated at the Bluetooth receiving end based on the monaural signal, so that the immersion of the user is enhanced.

Description

Technical Field

[0001] The present application belongs to the field of Bluetooth audio technology, and in particular, relates to an efficient Bluetooth receiver mono upmixing method, device, storage medium and equipment based on non-negative matrix decomposition. Background Art

[0002] The mainstream Bluetooth audio encoders are: SBC, AAC, aptx series, LDAC, LHDC, etc. After MP3, the most mainstream audio compression format is AAC, which has the characteristics of low bit rate and high sound quality and is widely used. Many music files are saved in AAC format, and many mid-to-high-end mobile phones and Bluetooth headsets support the use of AAC format to transmit music.

[0003] Bluetooth speakers are quite popular nowadays. In addition to enjoying mono and stereo sound, people hope to improve the sense of immersion. 5.1-channel surround sound is a widely used audio format, which can provide a better user experience. In multi-channel signals, one of the main functions of surround sound is to provide ambient sound effects, which is very important for immersion. However, for some reasons, a considerable number of AAC audio sources are mono. When these audio sources are played through multi-channel speakers, they can usually be played using the front center channel, or the front left plus front right channel. The speakers of the rear surround channels cannot be used, resulting in a waste of resources and failing to provide users with a sense of immersion. Summary of the invention

[0004] In response to the above-mentioned technical problems existing in the prior art, the present application provides an efficient Bluetooth receiving end mono upmixing method, device, storage medium and equipment based on non-negative matrix decomposition. Based on non-negative matrix decomposition, for the mono sound source transmitted in AAC format received by the Bluetooth receiving end, it is upmixed into the main sound source and ambient sound while being decoded at the Bluetooth receiving end, and then the Bluetooth receiving end plays the ambient sound while playing the main sound source, thereby enhancing the user's sense of immersion and improving the user experience.

[0005] In order to achieve the above-mentioned purpose, the first technical solution adopted in the present application is: to provide an efficient Bluetooth receiver mono upmixing method based on non-negative matrix decomposition, including: at the Bluetooth receiver, decoding the input mono AAC code stream or dual-channel code stream to time domain noise shaping to obtain spectral coefficients and their corresponding amplitude spectrum coefficients; performing mono decision on the spectral coefficients, if the decision result is mono, constructing the corresponding amplitude spectrum coefficient matrix according to the type of spectral coefficients; performing non-negative matrix decomposition on the amplitude spectrum coefficient matrix to obtain the corresponding basis matrix and activation matrix; calculating the main sound source spectral coefficients and the ambient sound spectral coefficients according to the spectral coefficients, amplitude spectrum coefficients, basis matrix and activation matrix; and continuing to perform the decoding process on the main sound source spectral coefficients and the ambient sound spectral coefficients respectively, and outputting the main sound source PCM data and the ambient sound PCM data.

[0006] The second technical solution adopted in the present application is: to provide an efficient Bluetooth receiving end mono upmixing device based on non-negative matrix decomposition, including: a module for decoding the input mono AAC code stream or dual-channel code stream to time domain noise shaping at the Bluetooth receiving end to obtain spectral coefficients and their corresponding amplitude spectrum coefficients; a module for performing mono decision on the spectral coefficients, and if the decision result is mono, constructing the corresponding amplitude spectrum coefficient matrix according to the type of the spectral coefficient; a module for performing non-negative matrix decomposition on the amplitude spectrum coefficient matrix to obtain the corresponding basis matrix and activation matrix; a module for calculating the main sound source spectral coefficients and the ambient sound spectral coefficients according to the spectral coefficients, the amplitude spectrum coefficients, the basis matrix and the activation matrix; a module for continuing to perform the decoding process on the main sound source spectral coefficients and the ambient sound spectral coefficients respectively, and outputting the main sound source PCM data and the ambient sound PCM data.

[0007] The third technical solution adopted in the present application is: providing a computer-readable storage medium storing computer instructions, wherein the computer instructions are operated to execute the efficient Bluetooth receiver mono upmixing method based on non-negative matrix decomposition in solution one.

[0008] The fourth technical solution adopted in the present application is: to provide a computer device, which includes a processor and a memory, the memory stores computer instructions, wherein the processor operates the computer instructions to execute the efficient Bluetooth receiver mono upmixing method based on non-negative matrix decomposition in solution one.

[0009] The beneficial effects that can be achieved by the technical solution of the present application are: the technical solution of the present application can be applied to both low-power Bluetooth and classic Bluetooth, and is based on non-negative matrix decomposition, for a mono sound source transmitted in AAC format received by a Bluetooth receiver, it is upmixed into a main sound source and ambient sound while being decoded at the Bluetooth receiver, and then the ambient sound is played while the main sound source is played at the Bluetooth receiver, providing the audience with an ambient sound experience, and keeping the position of the sound and image the same as that of standard multi-channel surround sound, thereby enhancing the user's sense of immersion and improving the user experience, and also utilizing the existing time-frequency transformation and overlap-addition to avoid increasing the algorithm delay. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 It is a flowchart of a specific implementation method of the efficient Bluetooth receiver mono upmixing method based on non-negative matrix decomposition of the present application;

[0011] Figure 2 It is a schematic diagram of a specific embodiment of the efficient mono upmixing of a Bluetooth receiver based on non-negative matrix decomposition of the present application;

[0012] Figure 3It is a schematic diagram of a specific implementation of the efficient Bluetooth receiver mono upmixing device based on non-negative matrix decomposition of the present application. DETAILED DESCRIPTION

[0013] The preferred embodiments of the present application are described in detail below in conjunction with the accompanying drawings so that the advantages and features of the present application can be more easily understood by those skilled in the art, thereby making a clearer and more definite definition of the protection scope of the present application.

[0014] A Bluetooth transmitter generally refers to a device with Bluetooth transmission function, such as a Bluetooth transmitter, Bluetooth transceiver, mobile phone, tablet computer, laptop computer, etc.

[0015] A Bluetooth receiver generally refers to a device with Bluetooth receiving function, such as a Bluetooth receiver, a Bluetooth transceiver, a Bluetooth speaker, or a device with a Bluetooth speaker, such as a multi-channel Bluetooth speaker, a car Bluetooth playback system, etc.

[0016] Nonnegative Matrix Factorization (NMF) makes all matrix components non-negative and achieves nonlinear dimensionality reduction. NMF has gradually become one of the most popular multidimensional data processing tools in research fields such as signal processing, biomedical engineering, pattern recognition, computer vision and image engineering.

[0017] The main sound source (i.e. direct sound) in multi-channel surround sound has the characteristics of repetitive structure and concentrated energy, while the ambient sound is relatively widely distributed in the spectrum. Since non-negative matrix decomposition can separate the repetitive structure and the area with concentrated energy in the data matrix, it can better estimate the main sound source, and then the ambient sound can be estimated by the difference between the original mono sound source and the main sound source.

[0018] Figure 1 It is a flowchart of a specific implementation method of the efficient Bluetooth receiver mono upmixing method based on non-negative matrix decomposition of the present application.

[0019] exist Figure 1 In a specific embodiment shown, the efficient non-negative matrix decomposition-based mono upmixing method for a Bluetooth receiver of the present application includes process S101, in which, at the Bluetooth receiver, the input mono AAC bitstream or dual-channel bitstream is decoded into time domain noise shaping to obtain spectral coefficients and their corresponding amplitude spectral coefficients.

[0020] In this specific implementation, at the Bluetooth receiving end, a mono AAC code stream or a dual-channel code stream is input, and decoding is performed to perform time domain noise shaping, and spectral coefficients are output. The dual-channel code stream includes dual-channel, stereo or joint stereo.

[0021] In a specific example of the present application, the mono AAC code stream or dual-channel code stream input by the Bluetooth receiving end can be generated by encoding the input mono audio signal at the Bluetooth transmitting end, and then the mono AAC code stream or dual-channel code stream generated at the Bluetooth transmitting end is sent to the Bluetooth receiving end, so that the Bluetooth receiving end receives the mono AAC code stream or dual-channel code stream. Normally, the input mono audio signal is encoded to generate a mono code stream; but the input mono audio signal can also be copied to form a dual-channel signal, and then encoded to generate a dual-channel code stream. In this case, although the code stream received at the Bluetooth receiving end is in a dual-channel format, its essence is still mono audio.

[0022] In a specific example of the present application, taking a sampling rate of 48 kHz as an example, there are two situations:

[0023] (1) Long window or transition window (Long start or Long stop) type:

[0024] The spectral coefficients are: Where N = 1024,

[0025] The amplitude spectrum coefficients are:

[0026] (2) Short window type:

[0027] The spectral coefficients are: Where N = 128,

[0028] The amplitude spectrum coefficients are:

[0029] In a specific example of the present application, if a dual-channel code stream is received at the Bluetooth receiving end, the above spectral coefficients are doubled, representing the spectral coefficients of channel 1 and channel 2 respectively.

[0030] exist Figure 1 In a specific embodiment shown, the efficient non-negative matrix decomposition-based mono upmixing method for a Bluetooth receiver of the present application includes process S102, performing mono decision on the spectral coefficients, and if the decision result is mono, constructing a corresponding amplitude spectrum coefficient matrix according to the type of the spectral coefficients.

[0031] Specifically, first, a mono decision is performed on the spectrum coefficients to determine whether it is mono.

[0032] In a specific embodiment of the present application, a mono decision is performed on the spectral coefficient, including: if the spectral coefficient is a mono spectral coefficient, the mono decision is skipped and mono is assumed by default; if the spectral coefficient is a binaural spectral coefficient and the spectral coefficients of the two channels are different, it is assumed to be binaural; if the spectral coefficient is a binaural spectral coefficient and the spectral coefficients of the two channels are the same, it is assumed to be mono.

[0033] In this specific embodiment, if the decoded spectral coefficients are mono spectral coefficients, this detection judgment can be skipped, and the current bitstream is mono by default; if the decoded spectral coefficients are binaural spectral coefficients, it is divided into the following two situations:

[0034] (1) The spectral coefficients of the two channels are different. If the spectral coefficients of the two channels are true binaural stereo spectral coefficients, then it is judged as binaural, which is not applicable to this application. Normal decoding and binaural stereo output can be used;

[0035] (2) If the spectral coefficients of the two channels are exactly the same, then they are actually mono channels, and this application is used to generate the main sound source and the ambient sound.

[0036] Taking the long window or transition window type as an example, the method of determining a single channel is briefly described as follows:

[0037] The spectral coefficients of the left channel are:

[0038] The spectral coefficients of the right channel are:

[0039] like The current code stream corresponds to a mono signal.

[0040] After the above-mentioned mono decision is executed, a decision result of whether it is mono is obtained. If the decision result is mono, a corresponding MDCT amplitude spectrum coefficient matrix is ​​constructed according to the type of spectrum coefficient.

[0041] In a specific embodiment of the present application, the types of spectral coefficients include long windows or transition windows, and short windows.

[0042] In this specific embodiment, the types of spectrum coefficients include two categories, one is a long window or a transition window, and the other is a short window. Among them, the long window and the transition window can be collectively referred to as a long window.

[0043] In a specific embodiment of the present application, a corresponding amplitude spectrum coefficient matrix is ​​constructed according to the type of spectrum coefficient, including: if the type of spectrum coefficient is a long window or a transition window, a long window matrix is ​​constructed; if the type of spectrum coefficient is a short window, 8 short window matrices are constructed.

[0044] In this specific embodiment, according to the two types of spectrum coefficients mentioned above, an amplitude spectrum coefficient matrix is ​​constructed:

[0045] (1) If the current frame is a long window or a transition window, a long window matrix V is constructed Long ,

[0046] (2) If the current frame is a short window, construct 8 short window matrices V Short,0 , V Short,1 ,···,V Short,7 .

[0047] exist Figure 1 In a specific embodiment shown, the efficient non-negative matrix decomposition-based mono upmixing method for a Bluetooth receiver of the present application includes a process S103 of performing non-negative matrix decomposition on an amplitude spectrum coefficient matrix to obtain a corresponding basis matrix and an activation matrix.

[0048] In this specific implementation, non-negative matrix decomposition is performed on the constructed amplitude spectrum coefficient matrix to obtain the corresponding basis matrix W and activation matrix H:

[0049] (1) Long window or transition window: Perform non-negative matrix factorization to obtain W Long and H Long ,

[0050] (2) Short window: Perform non-negative matrix factorization to obtain W Short,0 , W Short,1 ,···,W Short,7 and H Short,0 , H Short,1 ,···,H Short,7 .

[0051] For a non-negative matrix V, non-negative matrix decomposition is to find the optimal solution so that the product of W and H obtained by decomposition is an approximation of V. The steps of non-negative matrix decomposition are briefly described as follows:

[0052] (1) Construct the observation matrix V KL , which has K rows and L columns, where each column is the amplitude spectrum of a frame, and there are L columns in total;

[0053] (2) Initialize the basis matrix W KM ;

[0054] (3) Initialize the activation matrix H ML ; In a typical scenario, the above parameters can be taken as: long window: K=512, M=256, L=3~5, short window: K=64, M=32, L=3~5.

[0055] (4) Iterative solution based on KL divergence criterion is briefly described as follows:

[0056]

[0057] in, represents element-wise multiplication, T represents the transpose operation of the matrix, and I represents the identity matrix.

[0058] During the iteration process, the KL divergence of V and WH is calculated, and when it is less than a preset threshold, the iteration is stopped. In order to avoid the problem of abnormal conditions causing excessive calculation time, the number of iterations can also be preset, and when the preset number of iterations is reached, the iteration is stopped, thereby obtaining the basis matrix and the activation matrix.

[0059] exist Figure 1 In a specific embodiment shown, the efficient Bluetooth receiver mono upmixing method based on non-negative matrix decomposition of the present application includes process S104, which calculates the main sound source spectral coefficients and the ambient sound spectral coefficients according to the spectral coefficients, amplitude spectral coefficients, basis matrix and activation matrix.

[0060] Specifically, calculate the main sound source spectrum coefficient:

[0061] (1) Long window or transition window:

[0062]

[0063] (2) Short window:

[0064]

[0065] ······

[0066]

[0067] Calculate the ambient sound spectrum coefficient:

[0068] (1) Long window or transition window:

[0069]

[0070] (2) Short window:

[0071]

[0072] ······

[0073]

[0074] exist Figure 1 In a specific embodiment shown, the efficient non-negative matrix decomposition-based Bluetooth receiver mono upmixing method of the present application includes process S105, which continues to perform the decoding process on the main sound source spectral coefficients and the ambient sound spectral coefficients respectively, and outputs the main sound source PCM data and the ambient sound PCM data.

[0075] In a specific embodiment of the present application, the decoding process is continued to be performed on the main sound source spectral coefficients and the ambient sound spectral coefficients respectively, including: performing block switching and filter bank decoding processes on the main sound source spectral coefficients and the ambient sound spectral coefficients respectively.

[0076] In a specific embodiment of the present application, the efficient Bluetooth receiver mono upmixing method based on non-negative matrix decomposition of the present application also includes: using the front left channel, front center channel and front right channel of the Bluetooth receiver to play the main sound source PCM data.

[0077] In a specific embodiment of the present application, the efficient non-negative matrix decomposition-based mono upmixing method for a Bluetooth receiver of the present application also includes: performing decorrelation on the ambient sound PCM data to generate two-channel ambient sound signals; delaying the two-channel ambient sound signals for a predetermined period of time, and playing them using the rear left and right surround channels of the Bluetooth receiver.

[0078] In this specific embodiment, it is necessary to add a post-processing step to the ambient sound PCM data. First, decorrelation is performed (the decorrelation technology is a mature technology and is omitted here), and ambient sound signals of two channels are generated. Then, the ambient sound signals of the two channels are delayed for a predetermined period of time, wherein the predetermined delay is usually tens of milliseconds, and finally the rear surround sound channel is used for playback.

[0079] Figure 2 It is a schematic diagram of a specific embodiment of the present application's efficient mono upmixing at a Bluetooth receiver based on non-negative matrix decomposition.

[0080] like Figure 2 A specific embodiment shown, the present application's efficient Bluetooth receiver mono upmixing process based on non-negative matrix decomposition is: first, input AAC code stream, perform code stream encapsulation, Huffman decoding, inverse quantization, scaling, center / side stereo decoding, intensity stereo decoding to time domain noise shaping decoding, output spectrum coefficients, and obtain corresponding amplitude spectrum coefficients, then perform mono decision on the spectrum coefficients, for the signal with the decision result of mono, construct MDCT amplitude spectrum coefficient matrix, then perform non-negative matrix decomposition, and generate main sound source spectrum coefficients and ambient sound spectrum coefficients according to the result of non-negative matrix decomposition, then perform block switching and filter group on the main sound source spectrum coefficients and ambient sound spectrum coefficients, so as to output the main sound source (PCM) and ambient sound (PCM). This embodiment is based on non-negative matrix decomposition, for the mono sound source transmitted in AAC format received by the Bluetooth receiver, it is upmixed into the main sound source and ambient sound while decoding at the Bluetooth receiver, and then the ambient sound is played while the main sound source is played at the Bluetooth receiver, thereby enhancing the user's sense of immersion and improving the user experience.

[0081] In the efficient mono upmixing method for a Bluetooth receiver based on non-negative matrix decomposition of the present application, based on non-negative matrix decomposition, for a mono sound source transmitted in AAC format received by a Bluetooth receiver, it is upmixed into a main sound source and ambient sound while being decoded at the Bluetooth receiver, and then the ambient sound is played while the main sound source is played at the Bluetooth receiver, providing the audience with an ambient sound experience and keeping the position of the sound and image the same as that of standard multi-channel surround sound, thereby enhancing the user's sense of immersion and improving the user experience, and also utilizing the existing time-frequency transformation and overlap-addition to avoid increasing the algorithm delay.

[0082] Figure 3 It is a schematic diagram of a specific implementation of the efficient Bluetooth receiver mono upmixing device based on non-negative matrix decomposition of the present application.

[0083] exist Figure 3 In a specific embodiment shown, the efficient Bluetooth receiving end mono upmixing device based on non-negative matrix decomposition of the present application includes: a module 301 for decoding the input mono AAC code stream or dual-channel code stream to time domain noise shaping at the Bluetooth receiving end to obtain spectral coefficients and their corresponding amplitude spectrum coefficients; a module 302 for performing mono decision on the spectral coefficients, and if the decision result is mono, constructing the corresponding amplitude spectrum coefficient matrix according to the type of the spectral coefficient; a module 303 for performing non-negative matrix decomposition on the amplitude spectrum coefficient matrix to obtain the corresponding basis matrix and activation matrix; a module 304 for calculating the main sound source spectral coefficients and the ambient sound spectral coefficients according to the spectral coefficients, the amplitude spectrum coefficients, the basis matrix and the activation matrix; and a module 305 for continuing to perform the decoding process on the main sound source spectral coefficients and the ambient sound spectral coefficients respectively, and outputting the main sound source PCM data and the ambient sound PCM data.

[0084] In a specific embodiment of the present application, a mono decision is performed on the spectral coefficient, including: if the spectral coefficient is a mono spectral coefficient, the mono decision is skipped and mono is assumed by default; if the spectral coefficient is a binaural spectral coefficient and the spectral coefficients of the two channels are different, it is assumed to be binaural; if the spectral coefficient is a binaural spectral coefficient and the spectral coefficients of the two channels are the same, it is assumed to be mono.

[0085] In a specific embodiment of the present application, the types of spectral coefficients include long windows or transition windows, and short windows.

[0086] In a specific embodiment of the present application, a corresponding amplitude spectrum coefficient matrix is ​​constructed according to the type of spectrum coefficient, including: if the type of spectrum coefficient is a long window or a transition window, a long window matrix is ​​constructed; if the type of spectrum coefficient is a short window, 8 short window matrices are constructed.

[0087] In a specific embodiment of the present application, the decoding process is continued to be performed on the main sound source spectral coefficients and the ambient sound spectral coefficients respectively, including: performing block switching and filter bank decoding processes on the main sound source spectral coefficients and the ambient sound spectral coefficients respectively.

[0088] In a specific embodiment of the present application, the efficient Bluetooth receiving end mono upmixing device based on non-negative matrix decomposition of the present application also includes: a module for playing the main sound source PCM data using the front left channel, front center channel and front right channel of the Bluetooth receiving end.

[0089] In a specific embodiment of the present application, the efficient Bluetooth receiving end mono upmixing device based on non-negative matrix decomposition of the present application also includes: a module for performing decorrelation on the ambient sound PCM data to generate two-channel ambient sound signals; delaying the two-channel ambient sound signals for a predetermined length of time, and playing them using the rear left and right surround channels of the Bluetooth receiving end.

[0090] In the efficient Bluetooth receiver mono upmixing device based on non-negative matrix decomposition of the present application, based on non-negative matrix decomposition, for the mono sound source transmitted in AAC format received by the Bluetooth receiver, it is upmixed into the main sound source and the ambient sound while being decoded at the Bluetooth receiver, and then the ambient sound is played while the main sound source is played at the Bluetooth receiver, providing the audience with the experience of the ambient sound and keeping the position of the sound and image the same as that of the standard multi-channel surround sound, thereby enhancing the user's sense of immersion and improving the user experience, and also utilizing the existing time-frequency transformation and overlap-addition to avoid increasing the algorithm delay.

[0091] In a specific embodiment of the present application, a computer-readable storage medium stores computer instructions, which are operated to execute the efficient Bluetooth receiver mono upmixing method based on non-negative matrix decomposition described in any embodiment. The storage medium can be directly in hardware, in a software module executed by a processor, or in a combination of the two.

[0092] The software modules may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, removable disk, CD-ROM or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor so that the processor can read information from and write information to the storage medium.

[0093] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA), or other programmable logic devices, discrete gates or transistor logic, discrete hardware components, or any combination thereof. A general-purpose processor may be a microprocessor, but in an alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. The processor may also be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors combined with a DSP core, or any other such configuration. In an alternative, the storage medium may be integrated with the processor. The processor and the storage medium may reside in an ASIC. The ASIC may reside in a user terminal. In an alternative, the processor and the storage medium may reside in a user terminal as discrete components.

[0094] In a specific embodiment of the present application, a computer device includes a processor and a memory, wherein the memory stores computer instructions, wherein the processor operates the computer instructions to execute the efficient Bluetooth receiver mono upmixing method based on non-negative matrix decomposition described in any embodiment.

[0095] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0096] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0097] The above description is only an embodiment of the present application and does not limit the patent scope of the present application. Any equivalent structural transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. An efficient mono upmixing method for a Bluetooth receiver based on non-negative matrix decomposition, characterized in that: include: At the Bluetooth receiver, the input mono AAC bitstream or dual-channel bitstream is decoded to time-domain noise shaping to obtain spectrum coefficients and their corresponding amplitude spectrum coefficients; Performing a mono decision on the spectrum coefficient, and if the decision result is mono, constructing a corresponding amplitude spectrum coefficient matrix according to the type of the spectrum coefficient; Performing non-negative matrix decomposition on the amplitude spectrum coefficient matrix to obtain a corresponding basis matrix and an activation matrix; Calculate the main sound source spectral coefficient and the ambient sound spectral coefficient according to the spectral coefficient, the amplitude spectral coefficient, the base matrix and the activation matrix; as well as The decoding process is continued for the main sound source spectrum coefficient and the ambient sound spectrum coefficient respectively, and the main sound source PCM data and the ambient sound PCM data are output.

2. The efficient Bluetooth receiver mono upmixing method based on non-negative matrix decomposition according to claim 1, characterized in that: The performing monophonic decision on the spectral coefficients comprises: If the spectral coefficient is a mono spectral coefficient, the mono decision is skipped and the default is mono; If the spectral coefficients are binaural spectral coefficients, and the spectral coefficients of the two channels are different, it is determined to be binaural; If the spectral coefficients are binaural spectral coefficients, and the spectral coefficients of the two channels are the same, it is determined to be monophonic.

3. The efficient Bluetooth receiver mono upmixing method based on non-negative matrix decomposition according to claim 1, characterized in that: The types of the spectrum coefficients include long window, transition window and short window.

4. The efficient Bluetooth receiver mono upmixing method based on non-negative matrix decomposition according to claim 3, characterized in that: The step of constructing a corresponding amplitude spectrum coefficient matrix according to the type of the spectrum coefficient comprises: If the type of the spectral coefficient is the long window or the transition window, constructing a long window matrix; If the type of the spectral coefficient is the short window, 8 short window matrices are constructed.

5. The efficient Bluetooth receiver mono upmixing method based on non-negative matrix decomposition according to claim 1, characterized in that: Also includes: The main sound source PCM data is played using the front left channel, front center channel and front right channel of the Bluetooth receiving end.

6. The efficient Bluetooth receiver mono upmixing method based on non-negative matrix decomposition according to claim 1, characterized in that: Also includes: Performing decorrelation on the ambient sound PCM data to generate two-channel ambient sound signals; The ambient sound signals of the two channels are delayed for a predetermined time period, and played using the rear left and right surround channels of the Bluetooth receiving end.

7. The efficient Bluetooth receiver mono upmixing method based on non-negative matrix decomposition according to claim 1, characterized in that: The further decoding process is performed on the main sound source spectrum coefficient and the environmental sound spectrum coefficient respectively, comprising: Block switching and filter bank decoding processes are performed on the main sound source spectrum coefficients and the ambient sound spectrum coefficients respectively.

8. An efficient Bluetooth receiver mono upmixing device based on non-negative matrix decomposition, characterized in that: include: A module for decoding an input mono AAC code stream or a dual-channel code stream into time domain noise shaping at a Bluetooth receiving end to obtain spectrum coefficients and their corresponding amplitude spectrum coefficients; A module for performing a mono decision on the spectral coefficients, and if the decision result is mono, constructing a corresponding amplitude spectrum coefficient matrix according to the type of the spectral coefficients; A module for performing non-negative matrix decomposition on the amplitude spectrum coefficient matrix to obtain a corresponding basis matrix and an activation matrix; A module for calculating the main sound source spectral coefficient and the ambient sound spectral coefficient according to the spectral coefficient, the amplitude spectral coefficient, the base matrix and the activation matrix; A module for continuing to perform a decoding process on the main sound source spectrum coefficient and the ambient sound spectrum coefficient respectively, and outputting a main sound source PCM data and an ambient sound PCM data.

9. A computer-readable storage medium storing computer instructions, wherein the computer instructions are operated to execute the efficient non-negative matrix decomposition-based Bluetooth receiver mono upmixing method according to any one of claims 1 to 7.

10. A computer device comprising a processor and a memory, wherein the memory stores computer instructions, wherein the processor operates the computer instructions to execute the efficient Bluetooth receiver mono upmixing method based on non-negative matrix decomposition according to any one of claims 1-7.

Images